Detection of Covid-19 Cases Using Gan (Generative Adversarial Network)

Abstract

Novel Coronavirus disease (COVID-19) has abruptly and undoubtedly changed the world as we know it at the end of the 2nd decade of the 21st century. It is extremely contagious and its rapid growth has drawn attention towards its early diagnosis. Early diagnosis of COVID-19 enables health care professionals and government authorities to break the chain of transition and flatten the epidemic curve. With the number of cases accelerating across the developed world, COVID-19 induced Viral Pneumonia cases is a big challenge. Overlapping of COVID-19 cases with Viral Pneumonia and other lung infections with limited dataset and long training hours is a serious problem to cater. Limited amount of data often results in over-fitting models and due to this reason, model does not predict generalized results. To fill this gap, we proposed GAN based approach to synthesize Chest X-ray images which later fed in to deep learning models to classify images of COVID-19, Normal and Viral Pneumonia. Specifically, customized Wasserstein GAN is proposed to generate Chest X-ray images as compare to the real images. This expanded dataset is then used to train four proposed deep learning models: DeepCNN-CovNet19, ResNet-50, VGG-16, and GoogLeNet. The result showed that expanded dataset employed deep learning models to deliver high classification accuracies. In particular, DeepCNN-CovNet19 achieved 98% accuracy and proved to be the most efficient while VGG-16 achieved highest accuracy of 99.17% among all four proposed schemes. Rest of the models as ResNet-50 and GoogLeNet delivered 93.9% and 94.49% accuracies respectively. Later, the efficiency of these models is compared with the state of art models on the basis of accuracy. Our proposed models can be applied to address the issue of scant datasets for any problem of image analysis.

Note:
Funding Information: The research study did not receive any funding.

Declaration of Interests: The authors declare that there is no competing interest.

Keywords: COVID-19 Detection, generative adversarial network (GAN), Wasserstein GAN (WGAN), Deep Learning, Chest X-ray (CXR), Convolutional Neural Network(CNN), Residual Network(ResNet-50), Visual Geometry Group( VGG -16), Googlenet